Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
http://www.iaeme.com/IJCIET/index.asp 518 [email protected]
International Journal of Civil Engineering and Technology (IJCIET) Volume 8, Issue 11, November 2017, pp. 518–525, Article ID: IJCIET_08_11_055
Available online at http://http://www.iaeme.com/ijciet/issues.asp?JType=IJCIET&VType=8&IType=11
ISSN Print: 0976-6308 and ISSN Online: 0976-6316
© IAEME Publication Scopus Indexed
ACUTE LYMPHOBLASTIC LUKEMIA
DIAGNOSIS IN BLOOD MICROSCOPIC
IMAGES USING LOCAL BINARY PATTERN
AND SUPERVISED CLASSIFIER
A. N. Sruthi, S. Vinodhini, M. Shyamala Devi
Department of Computer Science and Engineering,
Veltech Dr. RR & Dr. SR University, Chennai, India
ABSTRACT
Acute lymphoblastic leukemia (ALL) is a subtype of acute leukemia that is well-
known among adults. The average age of someone with ALL is sixty five years. The
want for automation of leukemia detection arises when you consider that modern-day
methods contain manual exam of the blood smear as the first step towards prognosis.
that is time-ingesting, and its accuracy relies upon at the operator’s capacity. on this
paper, a easy method that mechanically detects and segments ALL in blood smears is
provided. The proposed method differs from others in: 1) the simplicity of the
advanced technique; 2) type of entire blood smear images in place of sub images; and
three) use of those algorithms to section and locate nucleated cells. pc simulation
concerned the following checks: comparing the effect of Haus Orff dimension on the
device earlier than and after the impact of neighborhood binary pattern, evaluating
the overall performance of the proposed algorithms on sub snap shots and entire pics,
and comparing the outcomes of a number of the existing structures with the proposed
machine. eighty microscopic blood photographs were tested, and the proposed
framework controlled to gain ninety eight% accuracy for the localization of the
lymphoblast cells and to separate it from the sub pics and whole pictures.
Keywords: Image Acquisition, Color Features and Color Correlation, Nuclei
Segmentation
Cite this Article: A. N. Sruthi, S. Vinodhini and M. Shyamala Devi, Acute
Lymphoblastic Lukemia Diagnosis in Blood Microscopic Images Using Local Binary
Pattern and Supervised Classifier, International Journal of Civil Engineering and
Technology, 8(11), 2017, pp. 518–525
http://www.iaeme.com/IJCIET/issues.asp?JType=IJCIET&VType=8&IType=11
A. N. Sruthi, S. Vinodhini and M. Shyamala Devi
http://www.iaeme.com/IJCIET/index.asp 519 [email protected]
1. INTRODUCTION
White blood cells (WBCs) or leukocytes play a chief role inside the diagnosis of different
sicknesses; as a result, extracting information about them is treasured for hematologists. The
time period leukemia comes from the Greek phrase “leukos” which means “white” and “goal”
which means “blood.” It refers to the most cancers of the blood or the bone marrow (wherein
blood cells are produced). Diagnosing leukemia is primarily based on the reality that white
cell count is accelerated with immature blast cells (lymphoid or myeloid), and neutrophils and
platelets are reduced. consequently, hematologists mechanically study blood smear below
microscope for proper identification and type of blast, cells. The presence of the excess
variety of blast cells in peripheral blood is a extensive symptom of leukemia.
ALL is frequently tough to diagnose considering the fact that the precise purpose of ALL
continues to be unknown. further, the signs of the disease are very much like flu or other not
unusual illnesses, consisting of fever, weak spot, tiredness, or aches in bones or joints. If the
described symptoms are present, blood assessments, inclusive of a full blood count number,
renal characteristic and electrolytes, and liver enzyme and blood count, have to be
accomplished. Given that there is no staging for ALL, selecting the type of treatment can
range from chemotherapy, radiation therapy, bone marrow transplant, and organic remedy.
Figure 1 suggests six different pics, three depicting wholesome cells from non-ALL patients
and 3 from ALL patients.
Figure 1 Images from ASH. (a)–(c) Myeloblasts from ALL patients. (d)–(f) Healthy cells from non-
ALL patients.
Otsu segmentation and automated histogram thresholding had been employed to section
WBCs from the blood smear picture. The paintings in hired contour signature to discover the
irregularities in the nucleus boundary. The work in employed selective filtering to phase
leukocytes from the opposite blood additives. This paper is dependent as follows. section II
focuses in detail on the system overview of the proposed version. Sections III and IV display
the photo processing techniques being used to perform enhancement and segmentation.
segment V builds the database of destiny vectors, which might be utilized by the type aspect
of the gadget. Sections VI and VII present the experimental outcomes of the classifier device
based on the capabilities extracted. section VIII consists of conclusions and future paintings
2. PPROCESS OVERVIEW
The gadget proposed guarantees step-by way of-step processing. Figure 2 depicts the gadget
review. The device review gives a detailed depiction of the series of steps which might be to
be followed for green class of acute leukemia. step one entails preprocessing the entire
pictures to conquer any heritage nonuniformity because of irregular illumination.
Acute Lymphoblastic Lukemia Diagnosis in Blood Microscopic Images Using Local Binary Pattern
and Supervised Classifier
http://www.iaeme.com/IJCIET/index.asp 520 [email protected]
Preprocessing also includes shade correlation wherein RGB snap shots are converted to L∗a∗b
colour area images. This step ensures perceptual uniformity. This step is observed by okay-
approach clustering to carry out the nucleus of every cell. Segmentation is followed by
function extraction primarily based on which classification and validation are carried out.
Figure 2 Process Overview
2.1. Preprocessing
CIELAB coloration functions and colour Correlation: The photos generated by way of
digital microscopes are generally in RGB colour area that is tough to segment. In practice, the
blood cells and picture heritage varies greatly with appreciate to colour and intensity. This
will be due to multiple motives along with camera settings, various illuminations, and ageing
stain. With a purpose to make the cellular segmentation sturdy with recognize to these
versions, an adaptive method is used: the RGB enter picture is converted into the CIELAB or,
extra effectively, the CIEL∗a∗b∗ colour area. Because of its perceptual uniformity, L∗a∗b
produces a proportional trade visually for a change of the equal quantity in shade value. This
guarantees that every minute distinction inside the shade cost is observed visually. Being
device unbiased is an delivered gain of the L∗a∗b colour area. Figure 3 presents the end result
of RGB to CIELAB shade conversion and segmentation procedure.
Figure 3 3RGB to CIELAB conversion and segmentation example.
A. N. Sruthi, S. Vinodhini and M. Shyamala Devi
http://www.iaeme.com/IJCIET/index.asp 521 [email protected]
2.2. Nuclei Segmentation
The aim of image segmentation is to extract vital records from an enter photograph. It
performs a key function since the performance of subsequent function extraction and type is
predicated significantly on the precise identification of the myeloblasts. Cluster evaluation
does now not use class labels that tag items with prior identifiers, i.e., elegance labels.
2.3. K-Means Clustering Algorithm
The k-approach set of rules calls for three consumer-distinctive parameters: the quantity of
clusters okay, cluster initialization, and distance metric. Every pixel of an object is assessed
into okay clusters based at the corresponding ∗a and ∗b values inside the L∗a∗b color space.
consequently, each pixel within the L∗a∗b color area is classed into any of the okay clusters
through calculating the Euclidean distance among the pixel and each colour indicator. these
clusters correspond to nucleus (high saturation), background (excessive luminance and low
saturation), and other cells (e. g., erythrocytes and leukocyte cytoplasm)
Figure 4 Feature set developed for the proposed system comprising of shape, color, texture features
2.4. Feature Extraction
Feature extraction in image processing is a technique of redefining a large set of redundant
data into a set of features of reduced dimension. Figure 5 illustrates the given algorithm. It
shows how the nucleus from a noncancer cell is superimposed with a grid of squares to
perform suitable box counting.
Figure 5 Superimposing of the nucleus with a grid of squares for box count measure.
LBP: The concept of local binary pattern (LBP) was introduced for texture classification.
The LBP method has proved to outperform many existing methods, including the linear
Acute Lymphoblastic Lukemia Diagnosis in Blood Microscopic Images Using Local Binary Pattern
and Supervised Classifier
http://www.iaeme.com/IJCIET/index.asp 522 [email protected]
discriminant analysis and the principal component analysis. In order to deal with textures at
different scales, the LBP operator was later extended to use neighborhoods of different sizes.
Shape Features with Their Illustrations
Shape features. One of the shape features that has proven to be a good measure for classifying
ALL by their shape is compactness. The shape of the nucleus, according to hematologists, is
an essential feature for discrimination of myeloblasts.
GLCM Features: Texture is defined as a function of the spatial variation in pixel
intensities. The GLCM and associated texture feature calculations are image analysis
techniques. Gray-level pixel distribution can be described by second-order statistics such as
the probability of two pixels having particular gray levels at particular spatial relationships.
This information can be depicted in 2-D gray-level cooccurrence matrices, which can be
computed for various distances and orientations. In order to use information contained in the
GLCM, Haralick defined some statistical measures to extract
Textual characteristic. Some of these features are the following.
• Energy: Also known as uniformity (or angular second moment), it is a measure of
homogeneity of image.
• Contrast: The contrast feature is a difference moment of the regional cooccurrence
matrix and is a measure of the contrast or the amount of local variations present in
an image.
• Entropy: This parameter measures the disorder of an image. When the image is not
texturally uniform, entropy is very large.
• Correlation: The correlation feature is a measure of regional-pattern linear
dependence in the image.
Color features. In addition to the features aforementioned, we have used the following
color-based feature.
1) Cell Energy: Also known as the measure of uniformity, it is the different Lab image
components. We define feature “δ” to be
Where represents the normalized GLCM element for the ith row and
jth column, and represents the ASM.
3. COMPUTER SIMULATION
The selection of a classification method for type is a challenging hassle due to the fact an
appropriate preference given the to be had records can drastically assist improving the
accuracy in credit score scoring practice. There’s a masses of statistical strategies, which
intention at solving binary class duties. In this paper, we use a aid vector system (SVM) for
constructing a selection surface in the function space that bisects the 2 classes, i.e., cancerous
and noncancerous, and maximizes the margin of separation among training of factors.
A. N. Sruthi, S. Vinodhini and M. Shyamala Devi
http://www.iaeme.com/IJCIET/index.asp 523 [email protected]
4. RESULTS AND DISCUSSION
Figure 6 (a) Image selection with pattern (b) Segmentation of pattern
Figure 7 (a) LBP Pattern (b) Classification
Figure 8 Segmentation and Clustering
Acute Lymphoblastic Lukemia Diagnosis in Blood Microscopic Images Using Local Binary Pattern
and Supervised Classifier
http://www.iaeme.com/IJCIET/index.asp 524 [email protected]
Figure 9 Clustering Results
5. DATASET ANALYSIS
For each image in the dataset, the classification/position of ALL lymphoblasts is provided by
expert oncologists. Furthermore, we suggest a specific set of figure of merits to be processed
in order to fairly compare different algorithms with the proposed dataset. The images of the
dataset have been captured with an optical laboratory microscope coupled with a Canon
PowerShot G5 camera. All images are in JPG format with 24 bit color depth, resolution 2592
x 1944. The ALL_IDB1 version 1.0 can be used both for testing segmentation capability of
algorithms, as well as the classification systems and image preprocessing methods. This
dataset is composed of 108 images collected during September, 2005. It contains about 39000
blood elements, where the lymphocyte has been labeled by expert oncologists. The images are
taken with different magnifications of the microscope ranging from 300 to 500.
6. CONCLUSION
In this paper, we've proposed a colour-primarily based clustering for segmenting white blood
cells from Acute Lymphoblastic Leukemia images. The technique efficiently combines RGB
and Lab shade space based unique-threshold methods to exploit their complementary
strengths. It consists of three main elements: preprocessing, segmentation, and post
processing. Historical past and red blood cells of the image are extracted through distinct
thresholds in segmentation procedure. The experimental outcomes advised that an overall
segmentation give high accuracy because the first step of computerized white blood cellular
differential system, it indicates a good prospect in further WBC function extraction, ALL
category, and diagnosis.
REFERENCES
[1] F. Scotti, Automatic morphological analysis for acute leukemia identification in peripheral
blood microscope images, in Proc. CIMSA, 2005, pp. 96–101.
[2] V. Piuri and F. Scotti, Morphological classification of blood leucocytes by microscope
images, in Proc. CIMSA, 2004, pp. 103–108
[3] M. Subrajeet, D. Patra, and S. Satpathy, Automated leukemia detection in blood
microscopic images using statistical texture analysis, in Proc. Int. Conf. Commun.
Comput. Security, 2011, pp. 184–187.
[4] H. Ramoser, V. Laurain, H. Bischof, and R. Ecker, Leukocyte segmentation and
classification in blood-smear images, in Proc. IEEE EMBS, 2006, pp. 3371–3374
A. N. Sruthi, S. Vinodhini and M. Shyamala Devi
http://www.iaeme.com/IJCIET/index.asp 525 [email protected]
[5] C. Reta, L. Altamirano, J. A. Gonzalez, R. Diaz, and J. S. Guichard, Segmentation of bone
marrow cell images for morphological classification of acute leukemia, in Proc. 23rd
FLAIRS, 2010, pp. 86–91.
[6] Abbas, N., et al., Nuclei segmentation of leukocytes in blood smear digital images. Pak. J.
Pharm. Sci, 2015. 28(5): p. 1801-1806.
[7] Mondal, P.K., et al., Segmentation of white blood cells using fuzzy C means segmentation
algorithm. IOSR Jornal of Computer Engineering, 2014. 1(16): p. 1-5.
[8] Rezatofighi, S.H., K. Khaksari, and H. SoltanianZadeh. Automatic recognition of five
types
[9] Dr. V. Goutham and M. Tejaswini, A Denial of Service Strategy To Orchestrate Stealthy
Attack Patterns In Cloud Computing, International Journal of Computer Engineering and
Technology, 7(3), 2016, pp. 179–186.
[10] P.S. Ardak, Dr. A.B. Borade and Renata Stasiak Betlejewska. An EPQ Model for
Deteriorating Items with Mix Demand Pattern. International Journal of Mechanical
Engineering and Technology, 8(6), 2017, pp. 59–69.